Profiling Python Packages

ddtrace

Datadog Python APM Client

py-spy

Sampling profiler for Python programs

gprof2dot

Converts profiling output to a dot graph.

viztracer

A debugging and profiling tool that can trace and visualize python code execution

austin-dist

Python frame stack sampler for CPython

scalene

Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals

fastapi-debug-toolbar

A debug toolbar for FastAPI.

scout-apm

ScoutAPM Python Agent. Supports Django, Flask, and many other frameworks.

easydev

Common utilities to ease the development of Python packages

timebudget

Stupidly-simple speed measurements for Python.

troncos

🪵 Troncos - Collection of logging, tracing and profiling tools

importtime-waterfall

Generate waterfalls from `-Ximporttime` tracing.

pelutils

The Swiss army knife of Python projects

django-query-counter

Count Django SQL queries and surface duplicate (N+1) and slow ones in the console — works as a decorator (management commands, tasks, views) or as middleware

calibrax

JAX benchmarking, profiling and evaluation metrics for Flax NNX: a registry of pure-function metrics (regression, classification, calibration, uncertainty, forecasting, generative, image, text, audio, graph, fairness), XLA FLOP counting, roofline analysis, GPU and energy monitoring, regression detection, publication exports, W&B and MLflow.

arena-pytest

A cross platform sandboxing framework with built in profiling workflows.

reprospect

A framework for reproducible prospecting of CUDA applications.

scitex-benchmark

Performance benchmarking, runtime monitoring, and profiling helpers

polars-telemetry

Profile Polars queries: per-node timings and row counts, both plans in a browser viewer, and findings on what slows a query down. Exports to OpenTelemetry and DogStatsD.

fastapi-profiler-lite

A lightweight, zero-configuration performance profiler for FastAPI applications

cutracer

A dynamic binary instrumentation tool for tracing and analyzing GPU kernel instructions.

prompt-flamegraph

Lightweight, zero-dependency Python package to profile LLM prompt tokens with interactive flamegraphs, waste detection, prompt diffs and HTML/SVG/Markdown/terminal exports.

executiontimer

Hierarchical execution timing for Python: nested sections, categories, asyncio and threads.

httptap

Rich-powered CLI that breaks each HTTP request into DNS, connect, TLS, wait, and transfer phases with waterfall timelines, compact summaries, or metrics-only output.